Back to skill

Security audit

EO Workflow: Paper

Security checks for vulnerabilities and agentic risk

Overview

This is a Markdown-only Chinese academic writing workflow with no install code, but users should verify sources and control any optional cross-session memory use.

Install only if you want a Chinese-language academic writing assistant. Treat generated literature claims, citations, plagiarism checks, and journal-format advice as drafts requiring independent verification, and avoid enabling EO memory for unpublished research details unless you understand how to inspect and delete stored data.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill description and usage guidance are entirely in Chinese, and there is no indication that users can select another language or that the skill is intentionally restricted to a Chinese-language audience. This may violate language-choice policy if the skill is expected to be generally usable.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The invocation examples are broad and can cause the skill to activate for loosely specified requests like 'complete a paper on [topic]' without clear boundaries, safety constraints, or scope limits. In a paper-writing workflow, this increases the chance of over-delegation into generating large amounts of academic content, including potentially misleading claims, fabricated citations, or undisclosed assistance, especially when the trigger language is open-ended.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill states it can remember advisor preferences and journal formats across sessions, but it does not provide a user-facing warning, consent mechanism, or explanation of what data is retained. Cross-session memory can expose sensitive academic information, including unpublished research directions, target journals, and supervisory preferences, creating privacy and confidentiality risks.

Content

No source excerpt is available for this finding.

Vague Triggers

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The description markets an end-to-end academic paper workflow but does not specify when the skill should or should not engage, making the operational scope ambiguous. This can lead to unintended use for ghostwriting or unsupported automation of research and publication tasks, which is risky in an academic context even if not directly malicious.

Content

No source excerpt is available for this finding.